1. Home
  2. Electronics Design
  3. Combining Enabled and LatchEnable pins to cascade decoders

Combining Enabled and LatchEnable pins to cascade decoders

I'm new to electronics and am responsible for developing the software for a data acquisition system. We have been working with an outside contractor for design guidance on the switching circuit but he is currently unavailable so I'm trying to pick up and run with the last few steps of the project.

The circuit is a straightforward signal switching design consisting of a couple of 74HCT4514 4>16 decoders and a 74HC238 to enable/disable them. The 74HCT4514 decoders are driving a series of CPC104 SSRs. A USB data acquisition module from measurement computing (great products) providers the interface for my application to control the data bits for the decoders.

Here is the relevant section:

formatting link

I have the circuit built up on a breadboard with SMT prototype boards for the SSRs and generally everything is working fine with one exception: The decoder is always enabled which means even when I pull all source pins down I'm still high on Y0. The problem is that only one SSR can be on at a time yet that's currently impossible (as I understand things) due to the fact the enable pin is hardwired to ground; as long as E is low then the output will be active. Again... I'm not an electronics guy (yet), so I could be way off but it looks like if I could invert the signal to E from LE I could achieve what I'm after. Basically, when the 74HC238 outputs H on Y0 it would pull down the E pin and pull up the LE pin on the first 74HCT4514. The same would happen on the second one when Y2 goes H and so on...

Here is a mockup of my proposed solution:

formatting link

I've added the logical inverter symbol but honestly I don't know if there is an IC that will do what I want in that schematic? I suspect I need a transistor that would switch on the signal to LE and when active would pull E down. I also can't leave E floating, it needs to be up or down.

So as you can see I'm in over my head and I would REALLY appreciate input from anyone who understands what I'm trying to accomplish or can maybe tell me if my proposed solution would work. At this point it's just an idea, I have no idea how to take it to the next step.

Thanks for reading, Steve

Reply to
steveklett
Loading thread data ...

--
There are many chips available that you can use to get the inversion
or you could use a single transistor as well.
Reply to
John Fields

Thanks for the help, John.

I've viewed your drawing and I don't understand what this section is: | 04 | |\ +--| >O |/

However it may be moot due to my original post being a bit misleading; My words and the schematic show only two 4514 but in reality there will be many, many more (~47 in total). I was trying to keep the example simple but it sort of backfired. Additionally there is the need to have all outputs from the 4514 low. I apologize for not being accurate/clear on my first post.

Reply to
steveklett

..

..

wire all /LE low and select with /EN ?

Reply to
langwadt 

--
That's a single inverter from a package of 6, a 74HC04.
Reply to
John Fields

e

of

d...

s
l

me

d...

I

an

.
,
g

I think I'd actually want to wire /LE high (to keep output selection alive) and bring /E low when I want output active. I think.... If this is true, then a 4515 (logic high) would maybe be a good choice... Thanks for the reply!

Reply to
steveklett 

--
I don't think that'll work; take a look at the truth table:

http://www.ti.com/lit/ds/symlink/cd74hc4514.pdf
Reply to
John Fields

f

...

e

...

I
n

Glad you don't mind :)

Outputs will be selected randomly. Actually it's funny... they will LITERALLY be selected randomly to as our system is designed to take sample readings from random locations. In my application I will write a 16bit value to the D/IO ports on the DAQ hardware which will enable one of the many outputs.

I read the datasheet again, specifically the first page where it talks about the behavior of /LE and /E and I think I can maybe accomplish what I want by pulling /LE high for ALL 4514s, then selectively pulling /E down for the 4514 I want active. I could accomplish this with 4515 (logic high). Your thoughts?

Reply to
steveklett 

--
Yes.
Reply to
John Fields

you are right, should have been /LE high

-Lasse

Reply to
langwadt

--- That'll work, except that a 4514 outputs high and a 4515 outputs low.

It also gets rid of the problem of more than output on at a time since, with /LE high, all the 4514 outputs go low.

If you have 48 4514's doing the output switching, then to select the chip you want to be active you could use 3 4515's, like this:

4515 +------+ |__ __| _ Vcc---------+-O|LE YO|O--->To U4 E | | . b0>-+-------|--|A0 . b1>-|-+-----|--|A1 . b2>-|-|-+---|--|A2 . b3>-|-|-|-+-|--|A3 . | | | | | |_ ___| _ b4>-|-|-|-|-|-O|E Y15|O--->To U19 E | | | | | +------+ | | | | | U1 | | | | | | | | | | 4515 | | | | | +------+ | | | | | |__ __| _ | | | | +-O|LE YO|O--->To U20 E | | | | | | . +-|-|-|-|--|A0 . | +-|-|-|--|A1 . | | +-|-|--|A2 . | | |-+-|--|A3 . | | | | | |_ ___| _ b5>-|-|-|-|-|-O|E Y15|O--->To U35 E | | | | | +------+ | | | | | U2 | | | | | | | | | | | | | | | 4515 | | | | | +------+ | | | | | |__ __| _ | | | | +-O|LE YO|O--->To U36 E | | | | | . +-|-|-|----|A0 . +-|-|----|A1 . +-|----|A2 . +----|A3 . |_ ___| _ b6>-----------O|E Y15|O--->To U51 E +------+ U3

b7> To A0, U4 -> U51 b8> To A1, U4 -> U51 b9> To A2, U4 -> U51 b10>To A3, U4 -> U51

So, with 11 bits out of the 16 you have available you can turn on any one of 768 outputs.

-- JF

Reply to
John Fields 

--
Oops... 

With /L high all the outputs go low.
Reply to
John Fields 

--
Aarghhhh!

With /E high all the outputs go low.
Reply to
John Fields

Hi John,

Excellent schematic, thanks for taking the time to do this. Your response is aligned with what I had been working on the last few days but was just starting to think may not be the "ideal solution" so seeing your version is great validation. I've purchased eagle CAD to create the final design and will be posting a link to my schematic on this thread in case you (or anyone else) is interested.

Now I just need to find the parts.... ;0)

Reply to
steveklett 

--
From Digi-Key:

http://search.digikey.com/scripts/DkSearch/dksus.dll?lang=en&site=US&KeyWords=74HC4514

http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=74HC4515
Reply to
John Fields

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.